The disclosure of Japanese Patent Application No. HEI 9-263595 filed on Sep. 29, 1997 including the specification, drawings and abstract is incorporated herein by reference in its entirety.
1. Field of the Invention
The present invention relates to the use of a CAD (Computer Aided Design) system for specifying information related to parts to be coupled at a weld point. More specifically, the invention is directed to the manipulation and joining of data to streamline the use of a CAD system for inputting and specifying such data.
2. Description of the Related Art
A CAD system is widely employed to design motor vehicles and various machines. CAD data constructed by the CAD system is widely used in production processes after the design has been completed.
In utilizing CAD data for a production process of welding, there are many cases where the constructed CAD data include what is called a xe2x80x9cweld point elementxe2x80x9d (hereinafter referred to simply as a weld point), which represents a point of spot welding. A welding robot or the like used in the production process retrieves necessary information imparted to the weld point in the CAD data and performs a welding operation using the information.
As described, information required to perform the welding operation is imparted to the weld point. The information includes a position of the weld point, an orientation of a weld surface (a normal vector thereof), a direction in which a welding gun of the welding robot enters, and the like. In addition, there is information describing a welding condition that depends on the material or total thickness of plate parts to be welded. This welding condition information must also be imparted. During known production processes, a welding robot carries out its welding operations using such information.
CAD data are also utilized in conducting structural analysis (an investigation on collision, strength, rigidity etc.) of a structural body such as a motor vehicle. In such a case, it has been proposed to utilize the information on the weld point as included in the CAD data. For example, the information on the position of the weld point, the orientation of the weld surface and the like as imparted to the weld point is used to couple those parts to be welded via the weld point, whereby a new finite element model is constructed. The CAD system automatically performs such a coupling operation, whereby it becomes possible to conduct structural analysis under the conditions that better match a real mechanical structure.
Various methods have been proposed to couple the parts. According to one method, nodes of the respective parts are displaced to the weld position to generate a shared node. According to another method, a node is assigned to the weld point and a finite element is provided to connect the node of the weld position with the nodes of the respective parts.
In this manner, there is proposed a method wherein the information imparted to the weld point included in the CAD data is used to automatically couple the parts via a predetermined finite element. However, this method encounters a case where the parts to be coupled cannot be selected accurately.
This is because the information on the coupling parts is not imparted to the weld point. In the CAD system according to the conventional art, there are many cases where the information on the parts to be coupled by means of welding is inputted as what is called an annotation.
On the other hand, various systems have been proposed to automatically couple those parts constituting a machine based on the information on a weld point. As has been described, however, the information on the coupling parts is not imparted to the weld point. Therefore, it is necessary to infer coupling parts based on the information on the given weld position. In general, the CAD system capable of automatically coupling parts in the thus-described manner widely adopts either a method wherein the nodes located in the vicinity of the weld position are automatically coupled via a finite element or a method wherein those nodes are coupled via a shared node.
According to a method of automatically coupling those nodes located close to the weld point, in the case where three parts A, B and C having flanges are superimposed on one another and only the parts B and C are welded to each other as shown in FIG. 9, it is difficult for the CAD system to accurately select parts to be coupled. FIG. 9 is a sectional view illustrating the angle part A, the plate part B and the angle part C that are superimposed on one another. Referring to FIG. 9, although the parts B and C are welded to each other at a weld point 10, the part A is welded to neither the part B nor the part C. In a machine of such a structure, if the conventional CAD system is used to automatically couple the nodes of the part models located close to the weld point 10 via a finite element, there is even a possibility of the part A, which is actually not supposed to be coupled to any other part, being coupled to some other part. Thus, there is raised a problem of inaccuracy in selecting coupling parts.
To prevent the part A from being automatically coupled via the finite element, it is desired that the weld point 10 have associated with it the appropriate information on the coupling parts. That is, the information that the parts B and C are coupled at the weld point 10 is newly imparted to the weld point 10. Thus, it becomes evident which parts are to be coupled at the weld point 10. Since it is clearly indicated that the parts B and C are to be automatically coupled, the part A is prevented from being automatically coupled by mistake.
However, a motor vehicle, for example, has thousands of weld points where welding should be carried out. Hence, if an attempt is made to provide all the weld points with the information on the coupling parts, the number of processes for inputting such information increases especially at a designing stage, which causes a problem of an increase in development costs. More specifically, the information on the coupling parts represents the part names of the coupling parts. That is, the part names of the parts B and C are imparted to the weld point 10. These part names have already been imparted to the finite elements that constitute the respective parts in the CAD system. For example, the constituents of the part B such as lines, faces etc. should be provided with the information that those constituents belong to the part B. Thus, the operation of inputting the part names has already been carried out with respect to the constituents such as lines, faces etc.
Accordingly, if the part names of the coupling parts are to be inputted to the weld point 10 again, it follows that the operation of inputting the part names such as the parts B and C is performed a plurality of times. Consequently, the inputting operation becomes considerably bothersome.
The term xe2x80x9cconstituentxe2x80x9d as used in the field of CAD corresponds to the term xe2x80x9cfinite elementxe2x80x9d in the field of CAE (Computer Aided Engineering). FIGS. 10A and 10B are explanatory views illustrating a relationship in terminology between CAD and CAE. Referring to those drawings, xe2x80x9cnodesxe2x80x9d and xe2x80x9csidesxe2x80x9d are called xe2x80x9cfinite elementsxe2x80x9d in the field of CAE (See FIG. 10A), whereas xe2x80x9clinesxe2x80x9d, xe2x80x9ccirclesxe2x80x9d, xe2x80x9cfacesxe2x80x9d, xe2x80x9carcsxe2x80x9d, xe2x80x9csplinesxe2x80x9d and the like are equivalent to the finite elements and called xe2x80x9cconstituentsxe2x80x9d in the field of CAD (See FIG. 10B).
When CAD data are used to conduct structural analysis, it is necessary to couple the parts of a finite element model based on the weld point. This is because such a coupling operation allows structural analysis to be conducted with precision. From the standpoint of model construction, it is efficient to simultaneously connect all the weld points via finite elements after preparation of part models corresponding to, for example, a motor vehicle. Thus, also from the standpoint of model construction, a method of inputting to a weld point the information on parts to be coupled at the weld point, such as part names, has been sought.
The present invention has been made in view of the aforementioned problem. It is thus an object of the present invention to provide a method of efficiently inputting to a weld point the names of parts to be coupled at the weld point in addition to a welding condition, and to provide a CAD system capable of conducting structural analysis with ease based on the inputted CAD data.
To solve the aforementioned problem, the present invention provides a method of constructing parts-coupling data for coupling at least two pieces of parts data to one another in a CAD system, including a coupling position designating step of designating a coupling position for the parts data and a coupling parts specifying step of specifying the parts data to be coupled at the coupling position designated in the coupling position designating step.
Since the aforementioned method includes the step of specifying the coupling parts, it is possible to couple the parts data, to which the information on the coupling parts have been imparted.
In addition, the method may include a finite element model constructing step of constructing a finite element model that represents a set of coupling parts obtained as a result of carrying out the specifying step one or more times. In this case, since the finite element model is automatically constructed, it is possible to make a calculation of strength or the like in accordance with an actual apparatus.
Another feature of the present invention is that the finite element model constructing step may include a step of connecting nodes of finite elements belonging to the coupling parts via a two-dimensional finite element such as a rigid-body element, a beam element or the like.
In this case, since only the parts to be coupled are connected via the two-dimensional finite element, it is possible to construct a finite element model with precision.
The finite element model constructing step may also include a step of connecting nodes of finite elements belonging to the coupling parts via a shared node. Also in this case, only the parts to be coupled are connected via the shared node. Thus, it is possible to construct a finite element model with precision.
Still another feature of the present invention is that the method may include a coordinate designating step of designating a coordinate of a coupling point representing the coupling position.
In this case, since the coordinate of the coupling point is designated, it is possible to accurately specify parts to be coupled based on the thus-designated coordinate.
Furthermore, the present invention also relates to a computer readable medium storing therein a program for realizing the aforementioned respective methods.
Furthermore, the present invention also relates to a system for realizing the aforementioned respective methods.